correlation-regime

Correlation-regime detection and crisis attribution — edge-density regime states with hysteresis, causal (no look-ahead) smoothing, regime-aware exposure context, first-mover crisis attribution with honest NAME / MACRO / AMBIGUOUS / ABSTAIN verdicts, and a correlation-rewiring leaderboard that catches slow bleed-outs

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./agent/src/skills/correlation-regime/SKILL.md

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with executable code, clear sequenced workflows, and strong validation discipline, but it is monolithic at ~500 lines with no progressive file-splitting and repeats its key caveats multiple times, holding conciseness and progressive disclosure at 2.

Suggestions

Split the four mode implementations into a scripts/ file (or reference modules) and keep SKILL.md as an overview that links one level deep, reducing the monolithic length.

De-duplicate the 'not a trade signal' caveat — state it once authoritatively and reference it from Mode 2, Notes, and the output template instead of restating it.

Consolidate the 'unpublished internal replays' disclaimer to a single place (the Overview already covers it) rather than repeating it in Mode 3 and References.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and does not explain basics Claude already knows, but at ~500 lines it repeats the same caveats several times — 'not a trade signal' appears in the Overview, Mode 2, Notes #2, and the output template, and the 'unpublished internal replays' disclaimer recurs in Overview, Mode 3, and References — so it could be tightened.	2 / 3
Actionability	Provides fully executable, complete Python functions with imports and docstrings (compute_edge_density, detect_regimes, first_mover_attribution, rewiring_leaderboard), a concrete dependency line, and a copy-paste output template — matching the score-3 'executable, copy-paste ready' anchor.	3 / 3
Workflow Clarity	Each mode has a numbered, sequenced workflow with explicit validation checkpoints — the Look-Ahead Warning ('re-check after replacing each filter with its causal version'), the Calibration Discipline walk-forward rules, and guard clauses that raise on bad thresholds/insufficient bars — plus an 8-item Notes checklist, matching the score-3 anchor.	3 / 3
Progressive Disclosure	Sections are well organized, but the skill is a single monolithic ~500-line SKILL.md with no bundle files (no references/, scripts/, or assets/ exist); the full code implementations and detailed methodology are inline rather than split into one-level-deep reference files, matching the score-2 'content that should be separate is inline' anchor.	2 / 3
	Total	10 / 12 Passed

Description

60%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and concrete about capabilities and uses correct third-person voice, but it lacks an explicit 'Use when...' trigger clause and leans heavily on jargon over natural user phrasing, capping completeness and trigger quality at 2.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks about correlation regimes, market fusing/diversification breakdown, who moved first in a crisis, or slow bleed-outs — not for static pairwise correlation snapshots.'

Soften the jargon in the lead clause ('edge-density regime states with hysteresis', 'NAME / MACRO / AMBIGUOUS / ABSTAIN verdicts') or move it after the natural-language triggers so the description reads as terms a user would actually say.

Add a one-line disambiguator against the sibling correlation-analysis skill (snapshot vs temporal) to reduce conflict risk.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — 'regime detection', 'crisis attribution', 'regime-aware exposure context', 'first-mover crisis attribution', and 'correlation-rewiring leaderboard' — matching the score-3 anchor of several specific actions rather than the partial score-2 anchor.	3 / 3
Completeness	It clearly states what the skill does, but there is no 'Use when...' clause or equivalent explicit trigger guidance, so per the judging guideline completeness is capped at 2 with 'when' only implied.	2 / 3
Trigger Term Quality	Contains some natural-ish terms a quant user might say ('correlation-regime', 'crisis attribution', 'slow bleed-outs') but they are buried under heavy jargon ('edge-density regime states with hysteresis', 'NAME / MACRO / AMBIGUOUS / ABSTAIN verdicts') and miss common natural variations like 'market fusing' or 'who moved first'.	2 / 3
Distinctiveness Conflict Risk	The temporal regime/attribution niche is somewhat specific and distinct from a snapshot, but it shares the correlation domain with a sibling 'correlation-analysis' skill referenced in the body, so a user mentioning correlation could overlap; no explicit triggers disambiguate.	2 / 3
	Total	9 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (505 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/src/skills/correlation-regime/SKILL.md
Commit: 8643fcd

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.